How to Appear in Google AI Overviews: A Practical Guide

How to appear in Google AI Overviews starts with a fact Google states in its own documentation. A page needs to be indexed and eligible to show a snippet, and nothing else is required. Google also says you need no AI-specific files or special markup.

Eligibility costs almost nothing, so selection is where the work sits. AI Overviews cite a handful of sources and answer the question on the results page itself. If an overview answers your buyer's research question and cites a competitor, that competitor gets the visit and the shortlist spot.

Google publishes no selection formula. Where the advice below rests on Google's documentation, we say so. Where it rests on patterns practitioners report, we flag that too.

How to appear in Google AI Overviews: the eligibility checks

To appear in Google AI Overviews, a page must be indexed and eligible to show a snippet in Google Search. Google lists no additional technical requirements. So confirm that Googlebot can reach the page and nothing blocks snippets. Then check that your key content sits in plain text.

Run five checks on every page you want cited. Most take minutes.

• Index status. Paste the URL into the URL Inspection tool in Search Console. If Google reports the page as not indexed, nothing else on this list matters yet.
• Crawl access beyond robots.txt. Google's documentation names CDN and hosting infrastructure next to robots.txt. A bot-protection rule at the CDN can block Googlebot while your robots.txt file looks fine.
• Snippet controls. A nosnippet tag, a data-nosnippet attribute or a tight max-snippet value limits what Google can show. Google names these as the controls for AI features. A rule added years ago for another reason can keep a page out.
• Text, not images. Google asks that important content be available in textual form. A pricing table saved as an image fails this check.
• Internal links. Google recommends making content findable through internal links. An orphan page gives Google fewer reasons to crawl it.

If the audit turns up problems on dozens of pages, that cleanup sits inside our AEO and generative engine optimisation work. A clean audit still guarantees nothing. Google states that meeting every requirement does not guarantee crawling, indexing or serving. That is how to appear in Google AI Overviews at the technical level, and selection comes next.

How query fan-out decides which pages get pulled in

Query fan-out is a technique Google says AI Overviews and AI Mode may use. Instead of running one search, the system issues several related searches across subtopics and data sources. It then builds one answer from the results. A page can earn a citation for a sub-question you never targeted, so anyone planning how to appear in Google AI Overviews should plan around sub-questions.

Google says this lets it show a wider and more diverse set of links than a classic search. That changes content planning. A page built around one keyword competes in one search. A page that answers six related sub-questions can compete in six.

Take a buyer asking how much SEO costs. A fan-out might split that into pricing models, price gaps between providers, retainer versus project fees, and payback time. Google does not publish its fan-out queries, so treat this split as an illustration. A page with a clear section for each sub-question gives the system four places to pull from.

Here is a working method:

1. List the follow-up questions. Write down the five to eight questions a buyer asks after the main one. The People Also Ask box and the related searches at the bottom of the results page are a fast source.
2. Give each question its own heading. Word the H2 or H3 the way a buyer would say it aloud.
3. Answer in the first two sentences. Add the detail after the answer, not before it.
4. Cut the rest. A page that tries to cover twenty sub-questions turns thin everywhere, so keep only the ones a real buyer asks.

Rank first, then make the passage easy to lift

Google says the usual SEO fundamentals still apply to AI features, and practitioners report that cited pages tend to rank well already. Because Google publishes no selection weights, treat ranking as the entry ticket and passage quality as the tiebreaker. How to appear in Google AI Overviews therefore splits into two jobs, and each needs a different fix.

If your page sits on page three, fix ranking before you rewrite a single sentence. That is the work behind our SEO services. Our guide on how long SEO takes to work will help you set realistic expectations inside your own team.

Google AI Overviews also work differently from chat assistants, because they sit on top of Google's own index. ChatGPT and similar tools follow other rules, so we cover them separately in how to rank in ChatGPT search. For a side-by-side view of where classic SEO ends and answer-focused work begins, read AEO vs SEO.

Write answer-first passages that earn AI Overview citations

AI Overview citations go to passages that answer a question on their own. Put a self-contained answer of roughly 40 to 60 words directly under a heading that matches the question. Then add the detail. The system can lift the short passage without needing the paragraphs around it to make sense.

The writing side of how to appear in Google AI Overviews comes down to the first sentence under each heading. Compare two openers for the same heading, "What is a canonical tag?"

Weak opener	Liftable opener
Canonical tags are one of those technical details many site owners wonder about, and there is a lot to unpack.	A canonical tag tells search engines which URL is the main version of a page when several URLs show the same content. Add it to the head of each duplicate page, pointing at the main URL.

The first version says nothing a reader could quote. The second names the thing, defines it and gives the action.

Four habits of a liftable passage

These four habits produce the second version:

• One question per heading. Use the wording a buyer types or says.
• Answer first, context second. The opening sentence should still make sense if a system quotes it alone.
• Name the subject. Starting an answer with "it" or "this" strips the passage of meaning once it leaves the page.
• Use a table or list for comparisons and sequences. Structured text is easier to extract than a long paragraph.

None of this guarantees a citation. Treat each rewrite as a test. Chat assistants reward the same answer-first structure, and we cover their citation habits in how to get cited by AI chatbots. Google says recrawling can take anywhere from several days to several months, so give each change time before you judge it.

What schema and llms.txt will not do for Google

Google says you need no special schema.org markup and no AI text file to appear in AI Overviews. Structured data is still worth adding when it matches the visible text on the page. Google lists no markup that triggers AI selection, so skip both as a shortcut for Google.

Several guides on how to appear in Google AI Overviews list schema markup and llms.txt as required steps. Google's documentation does not. Other AI products read the web differently, which is why we wrote separate guides on schema markup for AI search and llms.txt for SEO. Read them for those engines, and keep them out of the budget line for Google AI Overviews.

Which queries are worth chasing

Google says AI Overviews appear only when its systems judge them additive to classic Search, so they often do not trigger. That means many of your target queries may never show one. Before you spend time on how to appear in Google AI Overviews for a query, check that an overview shows for it.

Search the query in a signed-out browser window, then use this triage:

Overview shows?	You rank on page one?	What to do
Yes	Yes	Rewrite the key passages. This pairing has the highest return.
Yes	No	Fix ranking first, or write for a sub-question you can win.
No	Yes	Leave it alone. The classic result still earns the click.
No	No	Treat it as standard SEO work.

Then ask a money question. What is one click on this query worth to your pipeline? A research query that never converts does not deserve the same effort as a comparison query your buyers use right before they shortlist vendors.

How to measure Google AI Overviews optimization

Search Console now has a Generative AI performance report. Google announced it on June 3, 2026. It shows impressions in AI Overviews and AI Mode by page, country, device and date. It does not show clicks or queries, so pair it with your analytics data.

The measurement side of how to appear in Google AI Overviews has improved this year, but it still has gaps. Google began with a subset of sites, so check whether your property has the report. Three habits fill the gaps:

• Read impressions by page. The page view shows which URLs surface in AI features, so you can compare them against the pages you rewrote.
• Track conversions, not only sessions. Google says clicks from results pages with AI Overviews are higher quality, meaning people tend to spend more time on the site. Google has published no conversion data, so check your own.
• Run a monthly query audit. Search 20 priority queries in a signed-out window. Log whether an overview shows and who gets cited. Results vary by location and personalisation, so treat the log as a sample.

Where to start this week

Here is how to appear in Google AI Overviews, in order of priority:

1. Pick ten pages that already rank on page one for informational queries tied to revenue.
2. Run the five eligibility checks on each one.
3. List the follow-up questions and add the missing sections.
4. Rewrite the opening of each section as a standalone answer.
5. Record baseline impressions, then recheck after Google recrawls.

If you want help deciding which pages to fix first, talk to the AiM Growth team.